Thiols Decrease Human Interleukin (IL)

Thiols Decrease Human Interleukin (IL) 4 Production and IL-4-induced Immunoglobulin Synthesis By Pascale Jeannin,*Yves Delneste,* Sybille Lecoanet-Henchoz, * Jean-Francois Gauchat,* Paul Life,*$ Duncan Holmes,$ and Jean-Yves Bonnefoy*

From the *Glaxo Institutefor Molecular Biology, Immunology Department, Geneva, Switzerland; and $Glaxo Wellcome Research and Development Ltd ., Hertfordshire SG 1 2NY, United Kingdom

Summary N-Acetyl-L-cysteine (NAC) is an antioxidant precursor of intracellular glutathione (GSH), Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 usually given in humans as a mucolytic agent. In vitro, NAC and GSH have been shown to act on T cells by increasing interleukin (IL) 2 production, synthesis and turnover of IL-2 receptors, proliferation, cytotoxic properties, and resistance to apoptosis. We report here that NAC and GSH decrease in a dose-dependent manner human IL-4 production by stimulated peripheral blood T cells and by T helper (Th) 0- and Th2-like T cell clones. This effect was associated with a decrease in IL-4 messenger RNA transcription . In contrast, NAC and GSH had no effect on interferon y and increased IL-2 production and T cell proliferation . A functional consequence was the capacity ofNAC and GSH to selectively decrease in a dose-dependent manner IL-4-induced immunoglobulin (Ig) E and IgG4 production by human peripheral blood mononuclear cells. Interestingly, NAC and GSH also acted directly on purified tonsillar B cells by decreasing the mature e messenger RNA, hence decreasing IgE production. In contrast, IgA and IgM production were not affected. At the same time, B cell proliferation was increased in a dose-dependent manner. Not all antioxidants tested but only SH-bearing molecules mimicked these properties. Finally, when given orally to mice, NAC decreased both IgE and IgG1 antibody responses to ovalbumin. These results demonstrate that NAC, GSH, and other thiols may control the production of both the Th2-derived cytokine IL-4 and IL-4-induced Ig in vitro and in vivo.

Acetyl-L-cysteine (NAC) is a thiol antioxidant usu- volved in IL-2 production; NAC decreases nuclear factor ally used as a mucolytic drug (1), an antidote in acute KB (NF-KB) (11) and increases activator protein 1 tran- poisoning (2), and proposed as an anti-HIV agent (3) . NAC scription factors (12) . is also a precursor of intracellular glutathione (GSH) (4, 5). T cell responses have been divided in two subclasses, Thl GSH is found in millimolar concentrations in most eukary- and Th2, according to the profile oflymphokines produced, otic cells and plays important roles in several cellular func- IFN-y and IL-2 or IL-4, IL-5, and IL-13, respectively. A tions (free radical scavenging, conjugation of toxic sub- Thl response may favor cell-mediated immunity, whereas stances, and amino acid transport) (6) . Several studies have a Th2 response provides help for Immoral responses, in- indicated that NAC and GSH act on T cells by increasing cluding IgE and IgG4 (in humans) or IgGl (in mouse) iso- resistance to oxidative stress-mediated apoptosis (7), cyto- type switching. Some pathologies are associated with an toxic properties (4, 8, 9), synthesis and turnover ofIL-2 re- abnormal predominant Thl or Th2 response (13) . Thus, ceptors, IL-2 production, and proliferation (8, 10). NAC numerous studies are currently aiming at modulating lym- and some other thiol antioxidants have been reported to phokine production by T cells. In view of the effects of regulate the activation of some transcription factors in- NAC on T cells, we have evaluated whether NAC could selectively regulate Thl- or Th2-derived cytokine produc- 'Abbreviations used in this paper: DTT, dithiothreitol ; GSH, glutathione; tion. We report here that NAC and GSH decreased IL-4 GS-SG, oxidized GSH; mRNA, messenger RNA; NAC, N-acetyl-L- production by stimulated T cells and also acted on B cells cysteine; NFKB, nuclear factor KB ; SMC, S-methyl cysteine ; SOD, su- to selectively decrease IgE and IgG4 production. Both peroxide dismutase; TCC, T cell clones . these in vitro effects on T and B cells were associated with P. Jeannin and Y. Delneste contributed equally to this work. an action at the transcriptional level on IL-4 and IgE mes-

1785 J. Exp. Med. ©The Rockefeller University Press " 0022-1007/95/12/1785/08 $2 .00 Volume 182 December 1995 1785-1792

senger RNA (mRNA). In a mouse model, oral administra- lected and stored in aliquots at -70°C until tested for IgE, IgG4, tion ofNAC also resulted in a decrease of IgE and IgG1 re- IgM, and IgA by ELISA as described (17). In kinetic experiments, sponses to ovalbumin. NAC was added to stimulated B cells either at the beginning of the culture or at day 4 or 7. Proliferation Assays . Tonsillar B cells (2 X 105/200 p,1 per well), Materials and Methods T cells, and TCC (2 X 104/200 p,I per well) were cultured in 96- well cell culture plates and stimulated in for Compounds Tested. NAC, S-methyl cysteine (SMC), reduced triplicate either 3 d with IL-4 GSH and oxidized GSH (GS-SG), captopril, D-penicillamine, plus anti-CD40 mAb for B cells, or 48 h with PMA plus ionomycin or Con A for purified T cells and TCC, respec- L-methionine, L-camitine, ascorbic acid, superoxide dismutase tively, (SOD), and catalase were all purchased from Sigma Chemical Co. with or without different concentrations ofthe thiols tested. Cells were then pulsed for 6 h with [3H]thymidine (Amersham (St Louis, MO). Dithiothreitol (DTT) was purchased from Ald- International, Little rich (Buchs, Switzerland) . All the compounds were soluble in Chalfort, UK) before measurement of radio- activity. Results are expressed in counts per minute. culture medium, and the pH was adjusted to 7.4 before use. Molecular Analysis Isolation of Peripheral Blood T Cells, Characterization of T Cell of IL-4 and IgE Transcripts by Northern Blot . Tonsillar B cells were incubated for 8 d alone or in the presence Clones, and Isolation of Tonsillar B Cells. PBMC were isolated from ofIL-4, IL-4 plus anti-CD40 mAb, or IL-4 plus heparinized venous blood from healthy subjects by Ficoll-Hypaque anti-CD40 mAb plus (Pharmacia, Uppsala, Sweden), and T cells purified 10 mM NAC. T cells were stimulated for 6 h with PMA

were by ro- Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 plus ionomycin with or without 10 mM NAC Total was setting with SRBC. The purity of T cells, determined by flow . RNA isolated by extraction cytometry on a FACStar® Plus cell sorter (Becton Dickinson, Er- with RNAzol (Biotecx Laboratories Inc., Houston, TX) according to the manufacturer's instructions, sub- embodegem, Belgium) using FITC-labeled mouse anti-human jected to denaturing agarose electrophoresis (1 .5 CD3 mAb (Becton Dickinson), was >95%. The generation and p,g total RNA per lane), transferred to a nylon' membrane, and fixed by UV ir- characterization of the human CD4+ Th0 and Th2-like T cell radiation. Hybridizations with cRNA probes complementary clones (TCC) have been previously described (14, 15). T cells to were cultured in complete medium consisting of RPMI 1640 CE mRNA and IL-4 mRNA were performed as described on B and T cells, respectively (18) . (GIBCO BRL, Basel, Switzerland) supplemented with 10% heat- Flow inactivated human AB' serum (CTS, Annemasse, France), 2 mM Cytometric Analysis. After overnight incubation of purified human tonsillar B cells (2 X 105/200 p.l per well) with 200 glutamine, 20 mM sodium pyruvate (Sigma Chemical Co.), 50 U/ml IL-4 alone or in the presence of 0 g/ml streptomycin, and 50 U/ml penicillin (GIBCO BRL). .5-20 mM NAC, the sur- p, face phenotyping was performed by flow cytometry with FITC- Human B cells were isolated from Ficoll-separated tonsillar labeled mouse anti-human CD19 and anti-CD21 mononuclear cells by a two-step negative selection procedure us- mAbs (Becton Dickinson and Immunotech, Marseille, France, respectively) as ing sheep cell rosetting and magnetic bead depletion (Dynal, Oslo, described (17) . The mouse anti-human CD40 mAb (Serotec) Norway) to remove T cells. B cell purity was routinely >98% as determined by FRCS' analysis using FITC-labeled mouse anti- was detected using a FITC-labeled goat anti-mouse Ab (Bioart, Meudon, France). Control mouse IgG subclass Ab were all pur- CD20 and anti-CD3 mAbs (Becton Dickinson) . chased from Becton Dickinson. Induction of Lymphokine Production by T Cells. Purified T cells Modulation of Ovalbumin-induced IgE Production in Mice Treated were cultured in 96-well cell culture plates (Nuns, Roskilde, with NAC. Female 6-8-wk-old C57BL/6JIbm mice (Biological Denmark) (4 X 105/200 p,l per well) and stimulated with 10 ng/ Research Laboratories, Fullinsdorf, Switzerland) were injected in- ml PMA plus 1 p,M ionomycin (Calbiochem Corp., LaJolla, CA) . traperitoneally with 1 Kg ovalbumin (Sigma Chemical Co.; grade T cell clones (2 X 105/200 p,l per well) were stimulated with 10 IV) and 2 mg aluminium hydroxide (Serva, Heidelberg, Ger- g/ml Con A (Sigma Chemical Co.) or with immobilized anti- p, many) to favor specific IgE response (19) . Mice drank water con- CD3 mAb. The stimulations were set up in the absence or pres- taining 0.5, 1, or 2 g/liter NAC adjusted to pH 7.5 and changed ence of different concentrations of thiols. After 16 and 72 h of every 2 d. At different times after immunization, levels oftotal IgG culture, cell-free supernatants were collected and stored in ali- and ovalbumin-specific IgE, IgGl, IgG2b, and IgM Ab were mea- quots at -70°C. sured in sera by ELISA. Specific IgG1, IgG2b, and IgM were Quantification of Lymphokine in T Cell Supernatants . IL-2, IL-4, quantified using affinity-purified peroxidase-labeled Ab (The IL-5, and IFN-y were quantified by ELISA as follows: IL-2, us- Binding Site, Birmingham, UK; Biosys, Compiegne, France). ing two mAbs as previously described (sensitivity >200 pg/m1) Specific IgE was evaluated using a rat anti-mouse IgE mAb (20) (16); IFN-y, using a commercial kit (Genzyme Corp., Cambridge, revealed by a mouse peroxidase-labeled anti-rat IgG Ab Qackson MA; sensitivity >100 pg/ml) ; and IL-4 and IL-5, using mAbs ImmunoResearch Laboratories, Inc., West Grove, PA) . Sera were from PharMingen (San Diego, CA) (sensitivity >100 pg/ml) . diluted at 1:4 for specific IgE, 1:100 for specific IgGl, IgG2a, Lymphokines were quantified in the 16-h supernatants except for IgG2b, and IgM, and 1 :10,000 for total IgG quantification . IL-5 produced after stimulation with PMA plus ionomycin, which Statistical Analysis. Statistical analysis was performed using Stu- was measured in the 48-h supernatants . Results are expressed in dent's t test. nanograms per milliliter . Induction of IgE Production by Tonsillar B Cells and by PBMC. Tonsillar B cells and PBMC (2 X 105/200 p,l per well) were cul- Results and Discussion tured in 96-well culture plates in Iscove's medium enriched as Since NAC and GSH have been shown to increase pro- described (15). Purified B cells were stimulated with 200 U/ml liferation and IL-2 production by T cells (7, 8, 10), we in- IL-4 (Glaxo Institute, Geneva, Switzerland) and 0.1 p,g/ml anti- CD40 mAb (Serotec Kidlington, Oxford, UK). PBMC were in- vestigated whether they may modulate Thl/Th2 lympho- cubated with IL-4 alone or with IL-4 plus anti-CD40 mAb. Both kine production by human T cells. Stimulation ofperipheral these cultures were set up in the absence or presence of different blood T cells by PMA plus ionomycin or of TCC by anti- concentrations of thiols . After 14 d, cell-free supernatants were col- CD3 mAb or Con A was set up in the presence or absence 1786 Thiols Decrease IL-4 and IgE Production

of 0.5-20 mM NAC (Fig. 1) . Results showed that 0.5-20 (Fig. 2) . These data, reinforced by the fact that IL-4 was mM NAC decreased in a dose-dependent manner IL-4 quantified in the 16-h supernatants, allow us to exclude the production by purified T cells (Fig. 1 a) and by Th0- and possibility that the decrease ofIL-4 production could result Th2-like TCC (Fig. 1, b and c, respectively) stimulated for only from its consumption by proliferative T cells. NAC 16 h. In each case, a significant decrease was detected with and DTT have been shown to regulate the activation of 1 mM (P <0.05) and was maximal with 20 mM NAC. some transcription factors: They decrease NFKB and in- Similar observations were made with three other TCC (two crease activator protein 1 expression (11, 12) . One can spec- Th0 and one Th2 like) and with purified T cells stimulated ulate that thiols may affect the activation of transcription with anti-CD3 plus anti-CD28 mAbs (data not shown) . factors belonging to the NFkB family and implicated in the The decrease of IL-4 production was not due to a toxic ef- transcription of IL-4 (such as nuclear factor of activated T fect of NAC as assessed by trypan blue and propidium io- cells) (22). Further experiments are required to test this hy- dide staining of T cells stimulated in the presence of 20 pothesis. In parallel to the decrease ofIL-4 production, IL-2 mM NAC (data not shown). Moreover, in agreement with production was significantly increased by NAC up to 5 mM others (7, 8, 10), we found that 0.5-20 mM NAC increased (Fig. 1, a and b), as reported by others (10), whereas IFN-y in a dose-dependent manner the proliferation ofstimulated production was not affected whatever the time (Fig. 1, a T cells and TCC (Fig. 1 d) . GSH has been shown to in- and b) . NAC did not affect IL-5 production by T cells and Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 crease IL-4 internalization and degradation (21). Neverthe- Th0 TCC (Fig. 1, a and b) but decreased IL-5 production less, the decrease of IL-4 production induced by NAC was by the Th2-like TCC (Fig. 1 c). This decrease in IL-5 could associated with a decrease of IL-4 mRNA transcription be a consequence of the NAC-induced IL-4 reduction, as

Purified blood T cells Th0 (TT7)

E C

m C O i. U

0 0 0 .1 1 10 0 .1 1 10

Th2-like (POD)

Figure 1. E NAC decreases IL-4 production and in- a creases the proliferation of stimulated T cells. Purified T cells (a) and Th0 TCC (TT7) (b) were stimulated, w respectively, with PMA plus ionomycin or Con A in m the presence or absence of0.5-20 mM NAC. (a) IL-2 Y (X10) (A), O IL-4 (X0.03) (O), and IFN-y (O) were a. quantified in the 16-h supernatants by ELISA, and U IL-5 (O) in the 48-h supernatants. (b) Lymphokines were quantified in the 16-h cell-free supernatants'. (c) A Th2-like TCC (POD) was stimulated with Con A (top) or with anti-CD3 mAb (bottom) with or without 0.5-20 mM NAC. (d) Purified T cells (O) and a Th2-like TCC (POD) (0) were incubated, respectively, with PMA plus ionomycin or Con A in the presence of 0.5-20 mM NAC for 48 h and then pulsed for 6 h with [3H]thynudine . Results of five N-Acetyl-L-cysteine (mM) experiments ± SEM are presented .

1787 Jeannin et al.

Figure 2 . NAC decreases IL-4 brought about by addition of anti-CD40 mAb was de- rnRNA transcription in stimu- creased in a dose-dependent manner by NAC (IC S = 1 and lated T cells . T cells were incu- 2 mM, respectively) (Fig. 3 a) . A significant decrease of IgE bated for 6 h alone or with PMA plus ionomycin in the presence and IgG4 was observed with 0 .5 mM NAC (P <0 .05) and or absence of 10 mM NAC . was maximal with 20 mM . In contrast, IgA and IgM iso- RNA was isolated and used for types were not affected (Fig. 3 b) . Interestingly, NAC added Northern blot analysis using to highly purified B cells stimulated with anti-CD40 mAb probe complementary for IL-4 plus IL-4 also decreased to a higher extent IgE (IC_,,, = 0 .3 (top) . As a control, the Northern blot nylon membrane was nM) and IgG4 (IC50 = 0 .8 mM), but not IgA and IgM stained with methylene blue to (Fig . 3, c and d), while increasing B cell proliferation (Fig . 4 reveal rRNA (bottom) . a) . As observed for T cells, NAC was not toxic for B cells as assessed by trypan blue and propidium staining (data not shown) . Addition of NAC to IL-4 plus anti-CD40 mAb- previously reported (23) . These data indicate that NAC, stimulated B cells at days 0, 4, and 7 revealed that the stron- while increasing T cell proliferation, preferentially de- gest inhibitory effect ofNAC on IgE synthesis was observed creases IL-4, and under some circumstances IL-5, but has when NAC was added at day However, addition of 0 . NAC Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 no effect on IFN-y . at day 4 or 7 still led to a significant reduction in the syn- A key function of IL-4 is its ability to induce IgE and thesis of IgE (IC50 = 1 and 6 mM, respectively) (Fig . 4 b) . IgG4 production by human B cells (24, 25) . IgE produc- Thus, one can speculate that NAC may have therapeutic tion by B cells also requires a` physical interaction with T applications even on an ongoing IgE response . cells (26) or basophils and mast cells (27), involving a num- To investigate the potential mechanism ofaction ofNAC ber of surface molecules including CD40 (28) and CD21 on IgE synthesis, the expressions of IgE transcripts and some (29) . Based on the observation that NAC was able to de- key cell surface molecules, CD21 and CD40, were analyzed . crease IL-4 production by T cells, we tested if NAC would The mature E mRNA transcript was decreased by NAC in also inhibit IL-4-induced IgE and IgG4 production by B cells stimulated with IL-4 plus anti-CD40 mAb . Con- PBMC . NAC decreased both IgE and IgG4 production in versely, the sterile E mRNA transcript was increased (Fig . 5 a dose-dependent manner (IC 50 = 1 rnM) (P <0 .05) (data a) . NAC appears to be the first type of inhibitor ofIgE that not shown) . Even the superinduction of IgE and IgG4 can selectively decrease the mature E transcript. The other

PBMC B cells

Figure 3 . NAC decreases IgE and IgG4 production by human B cells. PBMC (a and b) and purified human tonsillar B cells (c and d) were incubated with IL-4 plus anti-CD40 mAb in the presence or absence of different concentrations of NAC . IgE (X 10 -1) (0), IgG4 (O), IgA (A), and IgM (+) production were quantified in the 14-d supernatants. Data from one out of three representative experiments are presented .

1788 Thiols Decrease IL-4 and IgE Production

reported IgE inhibitors, including IFN-y, IFN-o (30), and generation of IgE-producing B cells (31) . Taken together, TGF-(3 (18), inhibited both the mature and sterile e tran- these effects of NAC on IL-4 production, mature e scripts . These results indicate that NAC can act downstream mRNA transcription, and the expression of surface mole- of the IL-4-induced sterile e transcript by decreasing the cules involved in IgE regulation could at least partly explain generation of the mature e mRNA . Finally, the expression the decrease of human IgE production in vitro . of CD21 and CD40 was decreased in a dose-dependent manner by treatment ofpurified B cells stimulated with IL-4 in the presence of NAC (Fig. 5 b) . The expression of CD21 was totally inhibited with 20 mM NAC, whereas the expression of CD40 was only partly reduced (50% of decrease) . The expression ofother B cell markers, such as CD19 (Fig . 5 b), CD 11a, and CD20 (not shown), was unaffected . In fact, CD40-CD40 ligand interaction is essential for the switching process that takes place during the first 2 and 3 d ofthe B cell assays (Lecoanet-Henchoz, S ., unpublished results) .

The partial decrease of CD40 expression induced by NAC Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 cannot account for the total inhibition of the mature e mRNA expression . First, NAC still promotes B cell proliferation, although CD40 expression is decreased. Second, the ability of NAC added at day 7 to decrease IgE production cannot be explained only by a decrease of CD40 . The NAC-induced disappearance of CD21 expression could be responsible since the CD23-CD21 interaction favors the

a to- T

91 _b X

C G

Figure 5 . NAC decreases mature e mRNA, CD21, and CD40 expres- 0.1 1 10 sion on human tonsillar B cells . (a) Purified tonsillar B cells were incubated for 8 d alone or in the presence of IL-4, IL-4 plus anti-CD40 mAb, N-Acetyl-L-cysteine (mM) or IL-4 plus anti-CD40 mAb plus 10 mM NAC . RNA was isolated and used for Northern blot analysis using probes complementary to CE . The Figure 4 . NAC increases tonsillar B cell proliferation while decreasing lefthand upper panel shows an autoradiograph from a 12-h exposure IgE production in a time-dependent manner . (a) Tonsillar B cells were (alone ± IL-4) . The righthand upper panel shows a 1-h exposure (IL-4 + stimulated with IL-4 plus anti-CD40 mAb with or without 0 .1-20 mM otCD40 ± NAC) . As a control, the Northern blot nylon membrane was NAC for 3 d and then pulsed for 6 h with ( 3H]thymidine . (b) Tonsillar B stained with methylene blue (bottom) . (b) After overnight incubation o£ cells were stimulated with IL-4 plus anti-CD40 mAb and 1-20 MM NAC tonsillar B cells with IL-4 alone or in the presence of different concentra- added either at the beginning (O), at day 4 (O), or day 7 (A) of the cul- tions of NAC, the expression ofCD19 (O), CD21 (O), and CD40 (A) ture . IgE was quantified in the 14-d supernatants . a and b present data was evaluated by cytometry. Results from one (out of three) representa- from one (out of three) representative experiment. tive experiment are expressed in percentage of decrease of expression .

178 9 Jeannin et al .

These in vitro findings were extended to an in vivo mouse groups are critical for molecules to decrease IL-4 and IgE model. Mice were treated with ovalbumin to induce an IgE production. This was the case for NAC, GSH, DTT, and response and were given NAC dissolved in drinking water. also for captopril and D-penicillamine, two drugs used in 1 wk after immunization, both ovalbuniin-specific IgE and humans. In agreement with others, we found that GSH in- IgG1 (the mouse isotype equivalent to human IgG4) re- creased IL-2 production by T cells (data not shown) and T sponses were highly reduced (68 and 78% inhibition at 1 cell proliferation (Table 1) (7, 8, 10). Whereas GSH, capto- g/liter ofNAC, respectively) compared with untreated an- pril. and D-penicillamine used from 0.5 to 20 mM were as imals (Fig. 6) . A significant decrease was also observed us- potent as NAC in decreasing IL-4 and IgE synthesis, the ing 0.5 g/liter of NAC but not with lower concentrations dithiol DTT was the most efficient (ICs0 = 0.5 and 0.2 (data not shown). 2 wk after immunization, the inhibitory mM, respectively) . effect of NAC on IgE and IgG1 responses was still ob- NAC and GSH are able to increase IL-2 production and served (30 and 32% decrease, respectively). The synthesis of use T cell proliferation, and to decrease T cell apoptosis the other Ig isotypps was not significantly affected, in (7, 8, 10) . Because ofits ability to restore intra- and extra- agreement with the in vitro findings. On the basis of the cellular GSH levels and to decrease HIV replication (33, volume of water drunk in 1 d, the quantity of NAC swal- 34), NAC has been proposed in the treatment of HIV in- lowed could by a mouse be estimated at 50 mg/kg per day, fection (3, 5). It has been suggested that, during AIDS, a Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 comparable with 9 and 300 mg/kg per day given in hu- switch of the T cell cytokine pattern from a Thl to a mans as a mucolytic or as an antidote in poisoning, respec- Th2 type could be associated with a poor prognosis (35) . tively. This suggests that NAC acts in a range of doses that Moreover, IL-4-deficient mice have been shown to be re- are already used in human therapeutics. These results sug- sistant to retrovirus-induced immunodeficiency syndrome gest that in vivo NAC selectively down-regulates the pro- (36), and it has been very recently reported that NAC induction of antigen-specific IgE and IgG1 Ab, the two iso- duced a decrease ofIL--4 production by HIV-infected cells types regulated by mouse IL-4 (32) . (37). We report here that NAC decreases synthesis of the To investigate if antioxidants, in addition to NAC, simi- Th2-derived cytokine IL-4 and IL-4-induced Ig produc- larly reduce IL-4 and IgE production, we tested the com- tion while increasing B and T cell proliferation . Experi- pounds listed in Table 1. The antioxidants without an SH ments are in progress to define the molecular target(s) of group (such as L-carnitine, catalase, ascorbic acid, and SOD) thiols, as they may constitute new therapeutic molecules and the molecules without a free thiol (such as methionine, to treat Th2-mediated diseases such as allergic disorders SMC, and GS-SG) had no activity. It appears that free SH and AIDS.

anti-OVA anti-OVA 0 .5- 1 .257 anti-OVA IgE Ab IgG1 Ab IgM Ab 0 0 0.4-i 0 0 0.4- 0 1-

0.3- 0 0 0.3- o° 0.75 0 0 0.2-1 0 0 0 0 0.2-~ 0.57 0 o 0 0 0 0 0 0 0 0.1- 0 .1- 0.257

0 T- 0 0J t t t

0.5 anti-OVA anti-OVA Total IgG2a o IgG2b IgG 0.5 _~ 0 0.4- 0 0.6.6-- 0 0 0.4 0 8 0 0 .3- o g 0 g_ ... 0 .3 0 0 0.4- 0 0 Figure 6. Oral administration of NAC decreases 01 .21 0 .§0 specific IgE and IgGl Ab responses in ovalbumin- 0 0.2 0 0 immunized mice . C57BL/6JIbm mice injected in- 0.2- 0.1 traperitoneally with ovalbumin drank water con- 0.1- taining (+) or not (-) 1 g/liter NAC. 8 d after immunization, IgE, IgGI, IgG2a, IgG2b, and IgM 0 0 -'- I I 0 Abs directed against ovalbumn and polyclonal IgG + were measuredResults. are expressed in OD val- ues. Results from one set of experiments with nine N-Acetyl-L-cysteine (1 g/I) mice per group are presented.

1790 Thiols Decrease IL-4 and IgE Production

Table 1 . Thiol-containing Compounds Selectively Decrease IL-4 and IgE/IgG4 Production

Selective decrease Decrease in IL-4 Decrease in IFN-y Increase of of IgE/IgG4 production production proliferation production in vitro

Antioxidant free SH NAC + - + + GSH + - + + DTT + - + + Captopril + - + + D-penicillamine + - + + Antioxidant without SH L-camitine - - Not tested

Catalase - - - Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 Ascorbic acid - - Not tested SOD - - - Nonantioxidant SMC Methionine GS-SG

Summary of the effects ofNAC and other compounds on lymphokine production, proliferation, and IgE/IgG4 production by human lymphocytes, Three sets of compounds were tested: (a) antioxidants with a free thiol group (0.5-20 mM, NAC; 0.5-20 mM, reduced GSH; 0.1-2 mm, DTT; 0.5-20 mM, captopril; and 0.5-20 mM, D-penicillamine) ; (b) antioxidants without a free thiol group (0.5-20 mM, L-camitine; 50-400 U/ml, catalase; 10-100 p,M, ascorbic acid; 10-400 U/ml, SOD); and (c) S-substituted compounds (SMC, 0.1-20 mM; methionine, 0.1-20 mM; GS-SG, 0.1- 10 mM) . Purified peripheral blood T cells were stimulated with PMA plus ionomycin in the presence or absence of the compounds tested. Prolif- eration assays were then performed as described in Materials and Methods, and lymphokines were measured by ELISA in the 24- and 72-h cell-free supernatants . PBL synthesis of IL-4-dependent Ig was evaluated using the in vitro assay described in Materials and Methods in the presence or absence of the compounds tested.

We thank J.-P. Aubry for FACS® assistance, L. Potier for animal handling, and K, Hardy and J. Knowles for review ofthe manuscript and support.

Address correspondence to Dr, Jean-Yves Bonnefoy, Immunology Department, Glaxo Institute for Molecu- lar Biology, Chemin des Aulx 14, 1228 Plan-les-Ouates, Geneva, Switzerland.

Received forpublication 17 May 1995 and in revised form 18July 1995.

References Ventresca, G.P., V. Cicchetti, and V. Ferrari. 1989 . Acetyl- sponses by IL-2.j. Immunol . 152:5796-5805 . cysteine. In Drugs in Bronchial Mucology . P.C. Braga and L. 5. De Quay, B., R. Malinverrti, and B.H. Lauterburg . 1992. Glu- Allegra, editors. Raven Press, New York. 77-102. tathione depletion in HIV-infected patients: role of cysteine 2. Smilkstein, MJ., A.C. Bronstein, C . Linden, W.L. Augen- deficiency and effect oforal N-acetylcysteine. AIDS. 6:815- stein, K.W . Kulig, and B.H. Rumack . 1991. Acetaminophen 819. overdose : a 48-hour intravenous N-acetylcysteine treatment 6. Droge, W., K. Schulze-Osthofl; S . Mihm, D. Galter, H. protocol. Ann. Emerg. Med. 20:1058-1063. Schenk, H.P. Eck, S. Roth, and H . Gmunder. 1994. Func- 3. Roederer, M., S.W. Ela, F.J. Staal, and L.A. Herzenberg. 1992. tions ofglutathione and glutathione disulfide in immunology N-acetylcysteine : a new approach to anti-HIV therapy. AIDS and immunopathology. FASEBJ . 8:1131-1138. Res. Hum . Retroviruses . 8:209-217 . 7. Sandstrom, P.A., M.D. Mannie, and T.M. Buttke . 1994. In- 4. Yim, C.Y., J.B. Hibbs, J .R. McGregor, R.E. Galinsky, and hibition of activation-induced death in T cell hybridomas by W.E. Samlowski. 1994. Use o£ N-acetyl cysteine to increase thiol antioxidants : oxidative stress as a mediator ofapoptosis.J. intracellular glutathione during the induction of antitumor re- Leukocyte Biol . 55:221-226 .

179 1 Jeannin et al. 8. Liang, C.M., N. Lee, D. Cattell, and S.M. Liang. 1989. Glu- 245-248. tathione regulates interleukin-2 activity on cytotoxic T cells. J. 24. Lundgren, M., U. Persson, P. Larsson, C. Magnusson, C.I. Biol. Chem . 264:13519-13523 . Smith, L. Hammarston, and E. Severinson . 1989. Interleukin 9. Liang, S.M., C.M. Liang, M.E. Hargrove, and C.C. Ting. 1991 . 4 induces synthesis of IgE and IgG4 in human B cells . Eur. J. Regulation by glutathione ofthe effect oflymphokines on dif- Immunol. 19:1311-1315. ferentiation ofprimary activated lymphocytes. Influence ofglu- 25. Pene, J., F. Rousset, F. Briere, 1 . Chretien, J.Y. Bonnefoy, tathione on cytotoxic activity ofCD3-AK- ,J. Immunol. 146: H. Spits, T. Yokota, N. Arai, K. Arai, and J. Banchereau. 1909-1913. 1988. IgE production by normal human lymphocytes is in- 10. Eylar, E., C. Rivera-Quinones, C. Molina, I. Baez, F. Molina, duced by IL-4 and suppressed by a-interferon, y-interferon and C.M. Mercado. 1993 . N-acetylcysteine enhances T cell and prostaglandin E2. Proc . Natl . Acad . Sci. USA . 85:6880- functions and T cell growth in culture. Int. Immunol . 1 :97-101 . 6884. 11. Staal, FJ.T., M. Roederer, L.A. Herzenberg, and L.A. Herzen- 26. Vercelli, D., H.H. Jabara, K.I. Arai, and R.S. Geha. 1989. In- gerg. 1990. Intracellular thiols regulate activation of nuclear duction of human IgE synthesis requires interleukin 4 and factor KB and transcription of human immunodeficiency vi- T/B cell interactions involving the T cell receptor/CD23 rus. Proc. Natl. Acad. Sci . USA . 87:9943-9947. complex and MHC class 11 antigen. J. Exp . Med. 169:1295- 12 . Meyer, M., E. Schreck, and P.A. Baeuerle. 1993. H202 and 1307. antioxidants have opposite effects on activation of NF-KB and 27. Gauchat, J.F., S. Henchoz, G. Mazzei, J.P. Aubry, T. Brun-

AP-1 in intact cells: AP-1 as secondary antioxidant-respon- ner, H. Blasey, P. Life, D. Talabot, L. Flores-Romo, J. Thomp- Downloaded from http://rupress.org/jem/article-pdf/182/6/1785/1107341/1785.pdf by guest on 23 September 2021 sive factor. EMBO (Eur. Mol. Biol. Organ.)J. 12:2005-2015 . son, et al. 1993. Induction of human IgE synthesis in B cells 13. Romagnani, S. 1994. Lymphokine production by human T by mast cells and basophils. Nature (Loud.) . 365:340-343 . cells in disease states . Annu. Rev. Immunol. 12 :227-257 . 28. Jabara, H.H., S.M. Fu, R.H. Geha, and DJ. Vercelli. 1990. 14. Gauchat, J.F., J.P. Aubry, G. Mazzei, P. Life, T. Jomotte, G. CD40 and IgE: synergism between anti-CD40 monoclonal Elson, and J.Y. Bonnefoy. 1993. Human CD40-ligand: mo- antibody and interleukin 4 in the induction of IgE synthesis lecular cloning, cellular distribution and regulation of expres- by highly purified human B cells. J. Exp. Med . 172:1861- sion by factors controlling IgE production. FEBS (Fed. Eur. 1864. Biochem . Soc.) Lett. 315:259-266 . 29. Aubry, J.P, S. Pochon, P. Graber, K. Jansen, andJ.Y. Bonne- 15 . Life, P., J.P. Aubry, S. Estoppey, V. Schnuriger, and J.Y Bon- foy. 1992. CD21 is a ligand for CD23 and regulates IgE pro- nefoy. 1995. CD28 functions as an adhesion molecule and is duction. Nature (Loud.). 358:505-507 . involved in the regulation of human IgE synthesis. Eur. J. Im- 30. Gauchat, J.F., D.A. Lebman, R.L. Coffman, H. Gascan, and munol. 25:333-339 . J.E. De Vries. 1990. Structure and expression ofgermline ep- 16. Breittmayer, J.P., M. Ticchioni, B. Ferrua, A. Bernard, and C. silon transcripts in human B cells induced by interleukin 4 to Aussel. 1993. Ca(2+)-ATPase inhibitors induce interleukin-2 switch to IgE production. J. Exp . Med. 172:463-473 . synthesis and T cell proliferation. Cell. Immunol. 149:248-257 . 31. Henchoz, S., J.F. Gauchat, J.P. Aubry, P. Graber, S. Pochon, 17. Bonnefoy, J.Y., J. Shields, andJJ. Mermod. 1990. Inhibition andJ.Y. Bonnefoy. 1994. Stimulation ofhuman IgE produc- of human IL-4-induced IgE synthesis by a subset of anti- tion by a subset of anti-CD21 monoclonal antibodies: require- CD23/FCERII monoclonal antibodies. Eur. J. Immunol . 20: ment of a cosignal to modulate e transcripts. Immunology. 81 : 139-144. 285-290. 18 . Gauchat, J.F., H. Gascan, R. De Waal Malefyt, and J.E. De 32. Ohara, J., and W.E. Paul. 1987. Receptors for B-cell stimula- Vries. 1992. Regulation of germ-line epsilon transcription tory factor-1 expressed on cells ofhaematopoietic lineage. Na- and induction of epsilon switching in cloned EBV-trans- ture (Lond.). 325:537-540 . formed and malignant human B cell lines by cytokines and 33 . Roederer, M., F.J.T. Staal, P.A. Raju, S.W. Ela, and L.A. CD4+ T cells. J. Immunol. 148:2291-2299. Herzenberg. 1990. Cytokine-stimulated HIV replication is in- 19. Vaz, E.M., N.M. Vaz, and B.B. Levine . 1971. Persistent for- hibited by N-acetylcysteine . Proc. Natl. Acad . Sci. USA . 87: mation of reagins in mice injected with low doses of ovalbu- 4884-4888. min. Immunology. 21 :11-15 . 34. Mihm, S., J. Ennen, U. Pessera, R. Kurth, and W. Droge. 20. Baniyash, M., Z. Eshhar, and B. Rivnay. 1986. Relationship 1991. Inhibition of HIV replication and NF-kappa B activity between epitopes of IgE recognized by defined monoclonal by cysteine and cysteine derivatives. AIDS (Phila .). 5:497-503 . antibodies and the Fc epsilon receptor on basophils. J. Immu- 35 . Clerici, M., and G.M. Shearer. 1993. A THl -~ TH2 switch nol. 136:588-593 . is a critical step in the etiology of HIV infection. Immunol . 21 . Liang S.M., N. Lee, D.S. Finbloom, and C.M. Liang. 1992. Today. 14:107-111 . Regulation by glutathione of interleukin-4 activity on cyto- 36. Kanagawa, O., B.A. Vaupel, S. Gayama, G. Koehler, and M. toxic T cells. Immunology . 75:435-440 . Kopf 1993. Resistance of mice deficient in IL-4 to retrovi- 22. Szabo, SJ., J.S. Gold, T.L. Murphy, and K.M. Murphy. 1993. rus-induced immunodeficiency syndrome (MAIDS) . Science Identification ofcis-acting regulatory elements controlling in- (Wash. DC). 262:240-242 . terleukin-4 gene expression in T cells : roles for NF-Y and 37. Eylar, E.H., I. Baez, A. Vazquez, and Y. Yamamura. 1995 . NF-ATc. Mol . Cell. Biol. 13:4793-4805. N-Acetylcysteine (NAC) enhances interleukin-2 but sup- 23. Kopf, M., G. Le Gros, M. Bachmann, M.C. Lamers, H. presses interleukin-4 secretion from normal and HIV' CD4' Bluethmann, and G. Kohler. 1993. Disruption of the murine T-cells. Cell. Mol. Biol. (Oxf.). 41(Suppl . I) :S35-S40 . IL-4 gene blocks Th2 Cytokine responses. Nature (Loud.). 362:

1792 Thiols Decrease IL-4 and IgE Production